Glossary

Pharmacokinetic (PK) terminology used across the Estimation, Optimization, and Simulation screens.

Trough / Peak

The trough is the lowest drug concentration in a dosing interval, usually measured just before the next dose. The peak is the highest concentration, usually measured shortly after a dose is given (or after an infusion ends). Both are common sampling points when recording concentrations.

Cmin / Cmax / Cavg

The minimum, maximum, and average predicted concentration over a dosing interval, according to the model — as opposed to trough/peak, which are measured values, Cmin/Cmax/Cavg are figures the model computes from the fitted curve.

AUC (Area Under the Curve)

The total drug exposure over a time window — literally the area under the concentration-vs-time curve. Higher AUC means more cumulative exposure. AUCtrap and AUCgrap are two numerical methods (trapezoidal and using the model’s own solved curve, respectively) for calculating it.

MIC (Minimum Inhibitory Concentration)

For antimicrobial drugs, the lowest concentration that inhibits visible growth of a pathogen. Used as a reference point when setting optimization targets like time-over-MIC.

TOC (Time Over Concentration)

The proportion or duration of the dosing interval during which the concentration stays above a threshold (often the MIC) — a common target for antibiotics that work best when concentration remains elevated for a sustained period.

Clearance (CL)

The volume of blood/plasma the body clears of the drug per unit time. Higher clearance means the drug is eliminated faster.

Volume of distribution (VD)

A theoretical volume that relates the total amount of drug in the body to its concentration in plasma. It reflects how much the drug distributes into tissues versus staying in the bloodstream.

Intercompartmental clearance (Q)

In multi-compartment models, the rate at which drug moves between the central (blood) compartment and peripheral (tissue) compartments.

Typical value (TV-)

The population model’s expected parameter value for a “typical” patient (e.g. TVCL, TVVD, TVQ), before individualizing to this specific patient via estimation.

Steady state

The point in a repeated dosing regimen where the amount of drug entering the body per interval equals the amount being eliminated, so concentrations stop rising further with each dose. Optimization requires steady state.

Bayesian estimation

The statistical method PKPDrugs uses to combine the population model’s typical parameters with this patient’s own observed data (doses and concentrations), producing an individualized (“posterior”) parameter estimate that’s more precise than the population average alone.

Population vs. individual prediction

The population prediction is what the model expects for a typical patient with this patient’s covariates, before any of their own concentration data is used. The individual prediction is the Bayesian-updated curve specific to this patient, once their measured concentrations are taken into account.

Confidence interval (CI)

A range around an estimated value (e.g. “95% CI”) expressing the estimate’s uncertainty — a narrower interval means more confidence in the number.

Draft model

A new model file uploaded for an existing drug. Episodes keep using the model in force until an administrator publishes the draft. Administrators can test it on a real episode first.

Special character

Any character other than the letters A to Z and a to z and the digits 0 to 9 (an accented letter counts as a special character). Passwords must contain at least one.

Inactivity timeout

The time without activity (30 minutes by default) after which you are signed out.

Login lock

The wait after 5 failed sign-ins (the default): 1 minute, doubling with each further failure up to 60 minutes.